Re: arabic ISO characters

[Chahine Hamila <hchahine at magic dot fr>]

Salam guys,

Nadim Shaikli wrote:

> Salam all,

Salam Nadim,

> I'm getting pretty close to providing (as far as I can tell anyway) Arabic
> support in emacs-21 similar to what hebrew now gets, my only problem now is
> finding ISO-8859-6 bdf fonts and a means of writing something in that
> environment.

Nadim, I remember once getting my hands on a console fonts to bdf (or was it
pcf?) converter. I think I took it from Mark leisher site or something like
that, or it was one of his editor that could import and export to/from
different formats. The point is, I've created a very cool (I hope:)) Arabic
font for the text console that has position independant shapes. It is
delivered with Akka, and should exempt anyone working on Arabization for the
moment from taking care of the contextual formatting.

> You see I can use emacs' MULE fonts (pre-unicode) stuff which really doesn't
> make any sense and ought to, as far as I'm concerned, not be pursued at all
> (its a proprietary mapping that ought to be pretty much dead :-).

100% correct.

>  I think
> I've figured out how emacs (through leim and quail) gives one the ability to
> remap the keyboard to anything one chooses (for input support).  So now I'm
> at the point where I want to be able to enter Arabic letters (in unicode)
> and store them off for this remapping mechanism.  I would _really_ appreciate
> help here (any clues on what I can use to enter ISO-8859-6 characters -- I'm
> unable to complie akka

what's the problem?

> and had problems with axterm on solaris; using windows
> is a no go since its uses CP-1256)
>
> So in their hebrew file they show the following for ISO-8859-8,
>
>  ESC,HwESC(B
>
> for one of their letters (note the escape sequences), I need a similar
> method of getting all the Arabic characters mapped..

I am not sure I understand. What's that Hebrew letter you're talking about?
Sorry, I have xemacs-mule, but I'm too busy and lazy to consult:) What you're
doing (introducing Arabic to (x)emacs) is great. The only thing is, I don't see
the details of what you want to do, because I've never used any other input
than Latin one on xemacs. Don't worry about going technical, I know the basics
of elisp programming and have a rather detailed idea on how to remap keys on
(x)emacs.

> One other minor thing, which keyboard ought I follow -- what is the most
> common keyboard out there (can someone send me a mapping of the English
> letters to Arabic ones or a gif of the arabic keyboard).

I think there's no standard Arabic keyboard. In the Maghreb they use quite a
few different keyboards inside the same country, different constructors provide
different keyboards, Palestine has a Hebrew-Arabic-QWERTY keyboard, and I've
seen Saudi keyboards which were completely different from the ones I know. The
akka mapping is done on my (french) AZERTY-Arabic-Hebrew keyboard, and the
acon one is done on Ahmed's Egyptian one I guess, which I found out through
test that it was different from mine:)

Enjoy;) I think you'll have to find a way to make a generic Arabic wrapper that
would invoke a mapping from some keyboard mapping database.

> Mind you there will not be any shaping in this at all -- it will simply
> display Arabic characters (backwards no less - like how english is written).

Following the aunyxish bidi paper, it's not really a problem;)

The thing would be to find out a way to
 1) mirror a buffer display
 2) make the block cursor mode (make it the akka way? i.e. insert + backward
    movement every keystroke? dirty but trivial)
 3) Create the automatic input BC switch (this part should be the trickiest
    imho).

> The reason I think this is important - its simply to prove to the emacs guys
> that they _MUST_ support Arabic and that there is a framework in place now,
> so no excuses for them anymore and there are people out there capable of
> deductive reasoning and inspection (hey, I told you all that I'm no
> programmer :-)

heh:) You might already be doing more than most programmers;)